stanislaw_lemfandomcom-20200213-history
Informationbarrier
Information barrier? January 03, 1997, Stanislaw Lem There are complaints everywhere about the information flood. There is no way back. While some dream of intelligent agents, i.e. cunning servants who are supposed to take away the agony of choice, others shut themselves off. Or they allow themselves being flooded with senseless amounts of information. Stanislaw Lem considers whether the "megabit bomb" detonated will encounter not only human capacity limits, but also a technical barrier. In any case, the gradual replacement of human cognition by intelligent systems has one consequence - the possible early retirement of most people. In the book Summa Technologiae, which has now been in existence for a good thirty years, I introduced the metaphorical terms Megabit Bomb and Informationbarrier. The key to cognitive resources, I wrote at the time, is information. The number of scientists has increased rapidly since the industrial revolution. It has caused the well-known phenomenon that the amount of information that can be transmitted through one of the channels of science is limited. Science is a channel that connects civilization with the non-human and human world. The increase in the number of scientists means an expansion of the capacity of this channel. However, like any exponential growth, this process cannot continue indefinitely. If there is a shortage of candidates for science, the explosion of the megabit bomb will hit the information barrier. Limits of information processing Has anything changed in this picture after thirty years? First of all, I would like to note that several attempts were made to improve the absolute performance of the "final generation" of the computer. Developed according to the absolute speed known in physics, i.e. the speed of light, as is assumed today. However, the results of the estimates differed substantially. With the assumption of values that are specific to physics, i.e. utilizing the speed of light and the uncertainty relation according to Planck's constant, it was calculated that the most powerful computer that processes the data at the maximum achievable speed is a cube with an edge length of would be three centimetres. A premise unspoken with these assumptions, however, was an exclusively iterative way of calculation. This calculation, in its simplest form, characterizes the Turing automaton, which can only assume one of the two states: zero or one. You can perform any calculation of any linearly processing computer with the simplest Turing automaton. It's just that a Turing automaton takes an eternity to do what a Cray does in a split second. Parallel computers Quickly it was recognized that parallel processing computers could also be built. Although their programming and workings involve several tough problems to solve. What is in our skull proofs that such computers can be constructed. You see, the brain is mainly, though not exclusively, the real counterpart of a parallel computer, thanks to its design. It consists of two large hemispheres in which, surprising even to humans as constructors, are reigned by the unusual "strategy of allocation" of subordinate areas. For neurophysiologists, this was a real mess consisting entirely of puzzles. The researchers were able to identify symptoms of the failure of individual functions, e.g. with aphasia, amnesia, alexia and others. But they could not explain their functioning causally and functionally. Besides, as it should be, we still don't understand much of these and other phenomena in our brains. The brain can absorb information at a rate of 0.1 to 1 bit per second. Today, however, we are receiving a flow of new information at a rate that is between three and twenty bits per second. Information overload The entirety of human knowledge doubles approximately every five years, and this doubling time is continually decreasing. At the turn of the 19th and 20th centuries, this rate was still around 50 years. Every day, 7,000 articles get published around the world, and over 300 million magazines and 250,000 books get printed. On the other hand, there are already over 640 million radio and television sets. Since this information is already four years old, it is undoubtedly underestimating the actual number, mainly due to the vast increase of knowledge in the field of satellite television. The amount of information stored so far is said to be 10 to the power of 14 bits and will have doubled by the year 2000. Meanwhile, the brain's information acquisition capacity is already exhausted. Outside of science, it is much easier to notice the symptoms of this information asthenia than in science itself, especially if we limit its scope to the exact sciences. They are surrounded by a "halo" in the form of pseudo and quasi-sciences, which enjoy considerable popularity everywhere. As a rule, it is worthless and false "knowledge." Examples are astrology, botch-up, sectarian oddities of the "Christian Science" type and all "psychotronics" such as telepathy, telekinesis, "secret knowledge", news about the "flying saucers" or the "secrets of the pyramids". Those are pleasantly simple and fascinate with promises, like the one to clarify human fate or to understand the meaning of being alive. However, I am not restricting myself here to the area of exact science. It, incidentally, has long been flooded with cloudy counterfeits, which are not only harmful but also question the social status of science. Fraud is happening more and more in science, and it is encouraged by the still applicable rule; publish or perish. So many factors work together to increase information flooding. On the other hand, the above-mentioned unscientific sphere is subject to informative self-limitations, which a viewer with a satellite dish can quickly notice. In terms of content, almost all broadcasters worldwide differ very little from each other. That fact simply means that programs from different countries and in different languages are virtually the same in content. The weary audience is not concerned with deliberately "slimming down the television menu" or creating plagiarism. Simply, the viewers prefer to see variants of popular topics, which is why there are always new versions of any Tarzan or the Three Musketeers. In the US, it might be Fighting with the Indians or the Civil War. In Europe, however, topics would include the last world war there. The information allergy in the visual domain is particularly striking. The broadcasters fear innovation more than fire but value it as an innovative illusion. Of course, I am not playing a critic here. Because I do not want to evaluate and thereby lower the programs, which would only be superficial. I try to reveal the deeper, purely perceptual cause of this fact, which the television viewer knows from experience. The essence of television is a large quantity of supposedly different, independently produced programs that bear a strange resemblance to each other. When you turn off the sound, it is usually challenging to tell whether we are watching images from Turkey, Great Britain, the Netherlands, Sweden, Denmark or Spain. Because from everywhere, one receives almost the same semolina porridge with Pepper. The information overload dilemma An average person saves himself from information inundation both by reducing the bitstream and by eliminating what is somehow "not necessary" for mental absorption. In everyday life, this leads to increased ethnocentrism of the media, to a "growing thick-skinned" towards the content, which can shock or hurt feelings. However, science does not permit such a restraint. That being the reason for the growing weight of meals delivered to science by computer science with their host of computers. Like every new event, even if it is not that new anymore, computerization creates an indispensable sphere of life, which at the same time brings with it new worries. In countries where computerization has only just begun (which includes Poland), these concerns and this dilemma are not yet known. The first example explains where the catch can be. In the SF novel Return from the Stars (1960), I introduced Kalster to the plot as small devices that replace money transactions and circulation. There is indeed no place in a novel to describe the infrastructure of this "invention"! Today, however, magazines (e.g. in the United States) already write about smart cards based on this principle. There no longer has to be any money in circulation. Paying by check can also be a thing of the past now. Everyone has an account with a bank. And a smart card in their wallet. When paying, you give this card to the cashier, who puts it in the cash register, connected to the bank. The computer transmits to the bank computer how many currency units are to be deducted from the account. The same thing happens on the way payer (i.e. his computer) - the bank - the paid (i.e. his "Kalster"). All of that is wonderful - provided that no one can access our account using an electronic lock pick. As you know, computer crime and hackers have been around for a long time. They have even been able to get to the best-protected computers of various general staff. Cash can be buried or hidden in a treasury. However, the bank computers expose themselves to multiple attack venues, online or via radio. We are already so familiar with the phenomenon of viruses that it is not worth dealing with this "dark" side of computer science. Breaking the information barrier Computer networks, which, like the neurons in the brain, would be networked could serve to break through the "information barrier" in science. Each neuron, as we know, is indirectly or directly connected to a few tens of thousands of others. In short, the brain counts between 12 and 14 billion neurons. One understands this number wrong. It is about the number of connections and not the number of units that only work according to the flip-flop principle. On the other hand, computer giants would contribute to the breakthrough, for which my Golem XIV from the novel with the same title can stand. At this point, I should perhaps explain how I came across this golem and what the conception of an "over-Golem" growth of terabyte power can be, the growth of which is interrupted by periods of stagnation. This explanation is not a "pure fantasy" because I have always chosen the natural evolution of life on earth as my guiding star or rather as a guiding constellation. Perhaps the most peculiar phenomenon is the development of a succession of plant and animal species, which is characterized by a discontinuous increase in complexity. We only know of the first living beings that, within three billion years (at least, not at most), the genetic code was formed by them. This code shows an astonishing creative universality. From it came algae capable of photosynthesis, followed by bacteria, primeval animals, the molluscs, then fish, amphibians, reptiles and finally, mammals. They were crowned with the origin of the hominids and with humans at the top. However, vast abysses yawn between the species. For example, there were transitional forms between the reptile and the bird. Or between the fish and the amphibian. Yet, none of it remains. I considered these demarcations of genres, these zones of the genre silence, to be important. So important that I "transferred" them to a range of the culminations of the mind that followed them. They free themselves from the primary tasks given by physiology and anatomy, which the central nervous system of every living being (if it is an animal) has to fulfill. Paradoxes of information processing The idea that the largest constructible computer is a "cube" with an edge length of three centimetres must certainly be rejected. However, whether the construction of ever-larger computers will be a better way than creating a network based on the image of the neural networks of the brain can only be seen in the future. The comparison of the brain with the latest generation computer currently looks as follows; The brain is a parallel and multiprocessor system. It is composed of approximately 14 billion neurons that form a three-dimensional structure in which each neuron has up to 30,000 connections with other neurons. If a connection only performed one operation in a second, the brain would theoretically be able to perform ten trillion operations in that time. A neuron's flip-flop only lasts for a few milliseconds. The brain performs complex tasks such as recognizing and understanding language in about a second because it takes a few arithmetic operations. The computer, on the other hand, needs a million elementary steps for an analog task. Since a neuron cannot transmit complex symbols to a second neuron because it is a "simple" device of a flip-flop type, the performance of the brain depends on a large number of mutual neuronal connections. Thanks to these, we can easily use the language or languages. The task of multiplying two multi-digit numbers, on the other hand, is already a problem that not everyone can cope with. The phenomenon of incredibly skillful arithmetic artists, who can also be debilitated, is another mystery to me. Because it testifies to the existence of different subareas, which can even - or just - function more efficiently if other areas that are typically involved in the task are damaged! (Generally speaking, the brain tolerates damage much more easily than the computer). Today's supercomputers, as an expert claims, work at the developmental level of a five-year-old child (but it is about non-affective performance). It seems paradoxical that thousands of computers with the highest computing power would be needed to simulate the brain in real-time. In contrast, billions of people would need to do arithmetic calculations. An elementary operation of a neuron, as I said, takes about one millisecond. A computer, on the other hand, can do this within a nanosecond. So it works six times faster. Nevertheless, a person who comes into a café recognizes the face of his acquaintance in a split second, and the computer would take a few minutes to do this ... Automation of creation Probably the most important concern for our future, that is to say for the human future, seems to be the answer to the following question; Whether and how computers can create creative information capacity in the sense of authentic creation. The solution to an arbitrarily difficult mathematical problem has little in common with the work I am thinking of. Because the answer in the form of a solution is "hidden" in the mathematical structure of the problem. I take the liberty of returning to the book I quoted at the beginning. I wrote in it that the transition from non-renewable energy sources to new ones - from muscle power, the power of animals, the wind, water, coal or crude oil to atomic energy sources - requires information to be obtained beforehand. Only then, thanks to the trial-and-error method, the amount of information will exceed a "critical point." And the new technology based on it will open up new areas of energy and action for us. If the resources of fuel (coal, oil, gas), I had written, would have been used up by the end of the 19th century, it would have been doubtful whether we would have started atomic energy in the middle of the 20th century. Because its liberation required very large forces. They were first realized in the laboratories and then on an industrial scale. Nevertheless, as I wrote then, humankind is not at all ready (not even today) to switch to the exclusive exploitation of atomic energy ... The cold fusion announced by Fleischman and Pons, the cold fusion of deuterium with helium, was quickly dismissed as a mistake. However, recently, the Japanese, in particular, have started experiments in this area again. So "nothing is known for certain." I say this in the context of computer science. Because, with the definition of start parameters, which naturally arise from our current cosmological knowledge, we model the image of space in 100 billion years. This modelling has already been done on the computer, i.e. simulated, but one cannot draw any surprising insights from this simulation. That's because the start parameters lack any trace of them. Here is an example that may surprise some readers. Boleslaw Prus had Professor Geist , one of the characters in his novel Die Puppe , claim: "We have three cubes of the same size and made of the same material, but with different weights. And why? Because there are more steel particles in a full cube, less in the empty one, and the least in this one made of wire. Imagine that I managed to build cage-like particles instead of full particles and then you will understand the secret of the invention ... " And now a quote from an article in the scientific column of SPIEGEL (I do not quote the scientific press because I will be brief): "The scientists expect a 'completely new chemistry' that they hope for from the cage-like carbon spheres, called fullerenes. Created first in German laboratories, the number of their possible applications is vast. Fullerenes, named after Richard Fuller's self-supporting domed structures in architecture, are empty spheres that are formed from sixty atoms of coal. They connect to another like the connected pentagons that form a soccer ball. They can be used as construction elements in spaceships because they bounce off a steel plate undamaged when you fire at them at a speed of 27,000 km / h ... " The Japanese already synthesize cylindrical fullerenes that are filled with lead atoms and have drawn a wire with a thickness of a few atoms. The American press writes about Buckminsterfullerenes into which neon or helium atoms could be injected, and that's just the beginning. So what can you say about Boleslaw Prus's imagination? Over a period of one hundred years, it realized... It is a trivial thing that no one noticed this convergence. Because I believe, Prus himself did not believe in its realization very much. But I am not sure in this regard. I do not know why he wanted to write a novel entitled "The Fame" over the discovery made by Professor Geist and then gave up on this idea). The cage-like constructions (e.g. made of carbon atoms) were theoretically conceivable. One just didn't know how to carry out the synthesis - high temperatures and pressure are required. But this is information that must be given with the parameters of atoms in the initial assumption ... The computer still has no creative skills. However, I think that although the computer has not made the inventors and scientists unemployed, he has already started his invasion in another field. And its field - industrial production - poses the risk of mass unemployment. The future Half a century ago already, Norbert Wiener, the discoverer of cybernetics, predicted this unemployment in his book Human Use of Human Mind . This unemployment is caused by the increasing automation of more and more production processes. Who has not seen one of the many Japanese car factories on TV, where mostly giant yellow-enamelled robots whirl like crazy. And those robots in the absence of humans, assemble, weld, screw together the elements of bodies, engines, couplings. Such deserted factories are already emerging, and as some economists or engineers claim, unemployment will become an irreversible, increasing, and socially threatening phenomenon because the work of robotic machines is cheaper than human and often more accurate. These robots, whose world-army has already exceeded the number of three hundred thousand. They need no food, no wages, no rest and no vacation, no social security and also no social security in old age. In "old age" the robots go to the junkyard as wrecks). The robots' uprising or their rule, with which we have been frightened so often, is not threatening us. We simply face a bloodless conflict because the fruits of our minds and our civilization will simply make working people superfluous. Even if that is not happening today, it is not enough consolation, because the cost of investing in robots will continue to decrease. And as can be assumed, to them will belong to the 21st century. It seems that at the beginning of life, there was really the word: the word of the GENETIC CODE. And on the long road to the future, this word will give birth to the information that creates thoughtless and perfect workers from dead matter ... But what will our fate be then? That is a question that I cannot answer today. Source: https://www.heise.de/tp/features/Informationsbarriere-3563320.html